Integrative Biology 200A University of California, Berkeley "PRINCIPLES OF PHYLOGENETICS" Spring 2012 Lab 1: Introduction to PHYLIP What’s due at the end of lab, or next Tuesday in class: 1. Print out of Caminicules outfile and outtrees from pars 2. Print out of Azolla consensus output Setup 1. Have a laptop running either Windows or Mac. (Or you might get away with Linux, but I am not a Linux wizard.) 2. Download and install MESQUITE. (If this doesn’t work out of the box, you may have to download Java Virtual Machine as well, some older computers might not have it already.) 3. Download and install PHYLIP. Introduction Today we will be learning about some of the features of the PHYLIP (PHYLogeny Inference Package) software package. PHYLIP was developed by the famous evolutionary biologist Joe Felsenstein, works on most operating systems, and is available for free online. It is widely used, but slightly less popular than PAUP*. Like Mesquite, PHYLIP is an open source package, and you can make changes yourself if you program in C++. Methods available in the package include parsimony, distance matrix, and likelihood, as well as bootstrapping and consensus trees. Data types that can be handled include molecular sequences, gene frequencies, restriction sites, distance matrices, and discrete-state characters. What is PHYLIP? PHYLIP consists of about 30 programs that perform different algorithms on various types of data, and collectively are able to do most things you might want to do when it comes to inferring phylogenies. In PHYLIP, there is a program for each method you might want to use, and each type of data you might want to work with: Molecular sequence methods Program: What it does: protpars protein parsimony dnapars DNA sequence parsimony dnapenny DNA parsimony branch and bound dnamove interactive DNA parsimony dnacomp DNA compatibility dnaml DNA maximum likelihood dnamlk DNA maximum likelihood with clock proml Protein sequence maximum likelihood promlk Protein sequence maximum likelihood with clock dnainvar DNA invariants dnadist DNA distance protdist Protein sequence distance restdist Restriction sites and fragments distances restml Restriction sites maximum likelihood seqboot Bootstrapping/Jackknifing Distance matrix methods Program: What it does: fitch Fitch-Margoliash distance matrix method kitsch Fitch-Margoliash distance matrix with clock neighbor Neighbor-Joining and UPGMA method Gene frequencies and continuous characters Program: What it does: contml Maximum likelihood continuous characters and gene frequencies contrast Contrast methods gendist Genetic distance Discrete characters methods Program: What it does: pars Unordered multistate parsimony mix Mixed method parsimony penny Branch and bound mixed method parsimony move Interactive mixed method parsimony dollop Dollo and polymorphism parsimony dolpenny Dollo and polymorphism branch and bound parsimony dolmove Dollo and polymorphism interactive parsimony clique 0/1 characters compatibility method factor Character recoding program Tree drawing, consensus, tree editing, tree distances Program: What it does: drawgram Rooted tree drawing program drawtree Unrooted tree drawing program consense Consensus tree program treedist Tree distance program retree interactive tree rearrangement program Getting ready to use PHYLIP Input files: PHYLIP uses a standard format for its input. Here’s what it looks like, and what the parts mean: If you forget to extend the names to ten characters in length, using spaces or other characters, the program will not read the file in correctly and an error message will result. This is a rather persnickety format, but luckily, you don’t have to type it by hand. Mesquite can export this format for you. Go to File > Export… > Phylip (categorical data). A dialog will pop up. Set “Minimum length of taxon names” to 9 and “End of line character” to “Current System Default.” Then give your file a name that is both easy to remember and easy to type. Save it to the exe folder inside the Phylip folder. Step-by-step instructions for making a Phylip input file from scratch can be found at the end of this lab handout. [Step 1 – Make a PHYLIP input file] Last week, you and your lab partner made a matrix using Mesquite and the Caminalcules model data. You emailed it to me, so hopefully you have a record of it and you can download it now. Save it to the desktop and open it in Mesquite. File > Export… > Phylip (categorical data). A dialog will pop up. Set “Minimum length of taxon names” to 9 and “End of line character” to “Current System Default.” Then give your file a name that is both easy to remember and easy to type. Save it to the exe folder inside the Phylip folder. Using PHYLIP PHYLIP is not hard to use, but it is not intuitive either. You’ll need your “I’m comfortable with using command-line driven programs” hat. You’ll find PHYLIP folder under Start menu > Programs > phylip. (If you download PHYLIP yourself, you can place it in any folder—it is self-contained.) There are three files inside, doc, exe, and src. These contain documentation files, executables (ie, actual programs) and source code. You’ll want to keep all the files you’re working with in the exe folder, because to the best of my knowledge, PHYLIP can’t switch working directories. Each program in PHYLIP takes input in the form of a file, operates on the data, and produces output in the form of two new files. The figure below shows the default name for each of the files: ------------------- | | infile ---------> | | | | intree ---------> | | -----------> outfile | | weights --------> | program | -----------> outtree | | categories -----> | | -----------> plotfile | | fontfile -------> | | | | ------------------- This means if you include a file called infile in the exe folder, the program you are using will automatically read it in. If there is no file called infile, it will prompt you for the name of the file you want to use. Same goes for the output; all of the PHYLIP programs store their results in a file called outfile and a corresponding tree file called outtree. Phylip will warn you before overwriting them and give you a chance to name it something else. The easiest way to produce and read these files is probably with a text editor. Today we’re going to use a few different programs, so that you can see what PHYLIP can do. The first one is pars, which allows you to use unordered, multistate parsimony to infer phylogenies using discrete characters… the kind of characters that you made in lab last week. The second one is dnapars, which does the same thing, but for DNA data. We’ll also learn how to output tree drawings (drawtree and drawgram) and make consensues trees (consense.) pars [Step 2 – Analyze your Caminalcules file in pars] Ok, let’s experiment with pars. Open the pars program, start menu > phylip > exe > pars.exe. A window will pop up, and you’ll see the following: pars.exe: can’t find input file “infile” Please enter a new file name> Enter the name you gave your file. You’ll see the following: Discrete character parsimony algorithm, version 3.6 Setting for this run: U Search for best tree? Yes S Search option? More thorough search V Number of trees to save? 100 J Randomize input order of species? No. Use input order O Outgroup root? No, use as outgroup species 1 T Use Threshold parsimony? No, use ordinary parsimony W Sites weighted? No M Analyze multiple data sets? No I Input species interleaved? Yes 0 Terminal type (IBM PC, ANSI, none)? ANSI 1 Print out the data at start of run No 2 Print indications of progress of run Yes 3 Print out tree Yes 4 Print out steps in each site No 5 Print character at all nodes of tree No 6 Write out trees onto tree file? Yes Y to accept these or type the letter for one to change This is the PHYLIP menu, which allows you to se various options relating to the things the program can do. I’ll cover a few of the options, but you can check out the rest in the appendix at the end of this lab handout, and also the online documentation at http://evolution.genetics.washington.edu/phylip/phylip.html. (Be patient when reading the online documentation; there are a lot of places where you will be referred to other documents for the full story.) Today, we’re just going to change the Jumble option. First, we will randomize the input order of the species. This will allow pars to search a greater portion of tree of tree-space, giving us more confidence that we have found the most parsimonious result. Type j and hit enter. Enter any odd number to act as a random number seed. Enter 10 as the number of times to jumble (this will tell the program to randomly add the taxa together to form a starting tree 10 times; each of the 10 starting trees will then be used in its searches for a most parsimonious cladogram). Now hit Y. When the analysis is finished, the program window will close, and two new files, 'outfile' and 'outtree' will be added to your PHYLIP folder. You can open these files in Word or Text Edit to check the results. The outfile shows the cladogram(s) you have inferred, and gives some details of the results. The outtree gives you a simple, parenthetical representation of your cladogram(s). Rename these files to something that is easy to remember and easy to type. [Step 3 – Save these files so you can print them out to turn in later] drawtree and drawgram Now, you just created a tree from you Caminalcules data and I bet you want to look at it. PHYLIP offers you two ways to do this, drawgram, which draws rooted cladograms, and drawtree, which draws unrooted trees.